Abstinence from smoking produces aversive symptoms that prompt relapse, often within the first week following a quit attempt. Our prior work has characterized these nicotine abstinence symptoms in smoking cessation pharmacotherapy trials and identified those relating to smoking relapse. Extending this work, we propose to elucidate brain and behavioral mechanisms underlying medication effects on key early abstinence symptoms using arterial spin labeling (ASL) perfusion magnetic resonance imaging (MRI), and blood oxygen level dependent (BOLD) functional MRI (fMRI). Varenicline, an a4p2 nicotinic acetylcholine receptor partial agonist, will be used as a model medication, based on its superior clinical efficacy. The specific aims are: (1) to identify the neural substrates for varenicline effects on urges to smoke, using ASL perfusion MRI during a nicotine abstinent resting state, (2) To identify the neural substrates for varenicline effects on working memory during nicotine abstinence, using the N-back paradigm, and (3) To identify the neural substrates for varenicline effects on limbic reactivity to emotional stimuli during nicotine abstinence using a Face Emotion Identification fMRI paradigm. The study design is a within-subject double-blind crossover neuroimaging study of short-term (13 days) varenicline (vs. placebo) administration. Sixty treatmentseeking smokers will be scanned during 2 medication periods: (1) after 3 days of monitored abstinence while on varenicline, and (2) after 3 days of monitored abstinence while on placebo (medication order counterbalanced with a 2-week washout period). The primary outcomes are medication effects (within subject) on: (a) resting regional cerebral blood flow (rCBF) (ASL perfusion MRI), and (b) task-related BOLD activation (BOLD fMRI) after 3 days of abstinence. We will examine changes in behavioral performance and subjective symptoms in relation to brain activity changes. Following this investigation, all participants will be enrolled into a 12 week open-label varenicline trial, and we will explore associations of imaging data with clinical relapse. This research will contribute to pharmacotherapy development for nicotine dependence by: (a) providing a "neural fingerprint" against which future novel compounds can be compared, and (b) establishing the use neuroimaging as an early "surrogate marker" for medication response.